In the interest of improving cancer treatment, considerable attention has been placed on the modification of radiation damage. The major goal of this project continues to define and understand those aspects of unique tumor physiology, anatomy, cellular, and molecular processes that ultimately define the very nature of a tumor such that a particular dose of ionizing radiation, when used will be more effective. One means to that end is to investigate the interaction of ionizing radiation with a variety of chemotherapy agents to assess if tumors can be made more sensitive. We demonstrated that treatment of diverse human tumor cell lines with paclitaxel induced a block in G2/M of the cell cycle (radiosensitive phases of the cell cycle) which resulted in significant radiosensitization. As a consequence, translation of this pre-clinical information into a clinical trail combining radiation and paclitaxel in the treatment of head/neck cancer is currently underway in the Radiation Oncology Branch. Other chemotherapeutic agents (camptothecin derivatives, mitoxanthone, and UCN-01, among others) are being investigated to determine their use as radiation sensitizing agents. It is well established that hypoxia is a major determinant of radiation sensitivity. Therefore, we are using several murine tumor models to study tumor hypoxia. Our approach is to use current invasive techniques and extend that information to non- invasive methods that are under development, such that patient tumor treatment profiles may optimized on an individual basis. To that end we are using in vitro/in vivo survival techniques, fluorescent based oxygen probes, nitroimidazole fixation followed by immunohistochemistry, F-18 nitroimidazole PET scanning, blood oxygen level determination (BOLD) NMR, electron paramagnetic resonance, and Overhauser NMR. The information we acquire will teach us how to assess oxygen status in a tumor in a non-invasive manner to effect optimum treatment. Likewise, understanding and defining tumor vasculature and DNA expression should provide insight to understanding how neoangiogenesis impacts on tumor oxygen and consequently responsiveness to radiation, as well as provide information regarding those genes that are expressed in response to oxidative stress. Several angiostatin and endostatin expression vectors have been constructed for direct intra- tumor injection or systemic liposome delivery. Moreover, in vitro/in vivo tumor models containing vectors with switches for expression of endostatin and angiostatin are being made presently, such that the impact of vascularity can be assessed in the context of responsiveness to radiation and also in the context of non-invasive functional imaging. Gene expression within a tumor as a function of oxygen status, reperfusion, ionizing radiation exposure and general oxidative stress is being investigated to determine those molecular events that are important in the tumor cell responsiveness and survival to insult by ionizing radiation alone or in combination with pharmacologic agents using cDNA microarray technology. The aim is to extend our knowledge of tumor micro-environment, physiology, and molecular biology and to correlate such knowledge with non-invasive methods of studying tumors so that we design efficient and effective treatment protocols using radiation in combination or tandem with pharmacologic agents.